The predicted mine inflow in the ″Jixian Mining Area Geological Exploration Report″ is 250m3/h,but the present actual mine inflow is as high as 1700-2000 m3/h.

Forecasting the mine inflow is exactly important meaning for preventing mine water inrush, mine flooded and other malignant mine accidents.

To avoid the BP neural network common problem of trapped into a local solution and take advantage of the genetic algorithm\'s global optimization,a BP network optimized by genetic algorithm approach has proposed to predict mine inflow.

Chaos in mine discharge and its maximum predictable time scale;

Grey Markov Model for predicting mine discharge;

The spectral analysis method of the mine discharge's time series and its application;

The long-term evolvement of groundwater system in mining area is infected by the time series of mine hydraulic discharge.

On the basis of introducing the multifractals concept,power spectrum of mine hydraulic discharge act as a characteristic quantify is put forward,in order that the multifractal spectrum from time series of mine hydraulic discharge is calculated combining with Chhabra and Jensen put up instant calculating method.

On the basis of analyzing the relationship between fractional Brownian Motion and Joseph Effect, and discussing theoretical sense and practical calculable method in Hurst Exponent, the mine hydraulic discharge research using time seriess rescaled rangle analysis method is put forward.

By using the theory of grey control system,the mine water flow is analysedand predicated through establishing mathematical model in Yongchuan Mine,SichuanProvince.